Note that if two or more comparisons are planned, there are two kinds of type I errors: experiment-wise and comparison-wise error rates. The experiment-wise error rate is the probability of making at least one type I error when testing a whole collection of comparisons. The comparison-wise error rate is the probability of a type I error set by the analyst for evaluating each comparison.
State the type I error rate or level of significance (alpha-level) to be used for all analysis and whether statistical tests will be one- or two-sided. An example might be “All tests will be two-sided and considered statistically significant if P<0.05”.
Describe how you will adjust the level of significance when testing ...